Unlocking HPC Performance with AMD Tools and GPUs
A three-half-day online program for HPC scientists, researchers, and practitioners who want to port, optimize, and profile their code on AMD systems will be held from 4 to 6 November 2026.
The three half-day program is co-organized by AMD, NCC France (CC-FR), and CASTIEL3 within the EuroCC project. It combines introductory and advanced topics with hands-on exercises, helping participants learn to port, optimize, and profile scientific applications on AMD CPU and GPU systems. By pairing expert guidance with practical exercises, the program aims to equip participants with the knowledge and skills needed to use AMD CPUs and GPUs effectively for high-performance computing workloads.
Participants will learn to:
- Port applications from CPU or other vendor GPUs to run on AMD GPUs.
- Understand the ROCm SDK and what it provides
- What programming languages can be used for AMD GPUs (HIP, OpenMP, etc.)
- Optimizing and scaling GPU applications
- Communication packages including MPI, RCCL, and rocSHMEM
- Profiling and debugging tools for AMD GPUs